Skip to main content
Premium Trial:

Request an Annual Quote

Genoox Betting On Variant Classification Tech to Bring More NGS to Community Hospitals


CHICAGO (GenomeWeb) – Genomic analysis startup Genoox wants to accelerate the clinical uptake of next-generation sequencing by making its variant classification technology accessible to community hospitals and laboratories, not just academic medical centers.

The Tel Aviv, Israel-based company quietly entered the US market this year, then raised $6 million in a Series A round of private equity investment that closed in June. It currently has offices in New York and in Palo Alto, California.

Genoox has developed an artificially intelligent engine to classify genetic variants, based on American College of Medical Genetics and Genomics guidelines. This platform provides data analytics services to physicians and researchers looking to integrate genomic sequencing information into clinical workflows, and Genoox has said its technology can lower sequencing costs by as much as 90 percent and reduce turnaround time to a few minutes rather than hours or days.

"We're an enabling model to help hospitals, universities, and private [laboratory-developed testing] labs power their NGS programs," said Burt De Mill, director of US business development.

"Our true difference [from others] is that we are looking to be a true end-to-end supplier. I think many have claimed to be that, but I think our difference is the ability to go all the way to [a] clinical report that is patient-specific," De Mill said.

"When you're talking about genomes and exomes, this is real big data. I think our scalability and ability to load and analyze this massive amount of amount of data in minutes" gives the company an advantage," said CEO Amir Trabelsi, who cofounded Genoox in 2014.

"There are lots of folks out there doing annotation variant calling with high degrees of sensitivity and specificity, but I think where we're really focusing is further downstream in the bottleneck that we see, which is clinical reporting," De Mill added. "What does the clinician do with the patient [who's] staring him in the face right across the desk? What drug do you put me on? What do I do now?"

Addressing this bottleneck, according to De Mill, helps lower the barrier to entry for community hospitals and independent labs looking to get into clinical genomics.

"Now that we have variant calling and interpretation down pretty well," De Mill said, "I think the final frontier is clinical metadata — not just the variant itself, but the variant in context with the patient's clinical history."

Earlier this year, the Israeli Ministry of Health contracted with Genoox to analyze the genome sequences of more than 100,000 residents of that country. Israel has nationwide electronic health records, making it easier to get clinical records on just about anyone.

"I think where personalized medicine gets extremely interesting and extremely useful is when you can take metadata from disparate places, not just genomic data — but other clinical history data — and start making comparisons," De Mill said.

It's about being able to act on data right away for a patient sitting in front of a doctor. "That issue has yet to be fully solved, and I think that when that can be done in minutes — rather than a genetic counselor, quite frankly, sitting in front of a computer doing PubMed searches, which is what they do now — when that can be done with some AI, I think the speed and cost is going to drop dramatically," he said.

De Mill believes that this technology will help alleviate the shortage of geneticists and genetic counselors, De Mill added. "It's going to make their jobs much easier and it's going to make them much more efficient. Instead of dealing with five patients a day, they can deal with 50," he said.

The target market, particularly in the US, is community hospitals, not just academic medical centers. Although patients typically go to teaching hospitals for rare diseases that might involve genetic testing, like any advance, genomics is starting to move beyond academia into community hospitals and clinics.

"We think we have two models that could be really helpful," De Mill said. "One is to power academic centers to make their platforms more clinically useful and really give them an 'easy button' at the academic centers. But we also believe folks at Tier 2 hospitals out in the community are interested in this technology, and private LDT labs that are also developing their NGS programs could gain from having a ready-to-go, plug-and-play platform."

This is both clinical decision support and applied genomics. "Clearly, the variants are important, the pathogenicity of those variants, particularly in things like rare disease, and somatic mutations," De Mill said.

"Now that variant calling has really gotten up to 99, 99.5 percent accuracy with sensitivity and specificity, the next question is, for Patient X, who may be a female under the age of 30, premenopausal, no prior clinical history, how do I use that variant to try to treat them with the right drug or the right protocol based on their clinical history in their particular setting?"

Eventually, Genoox wants to be able to analyze germline mutations as well as somatic mutations. "We're looking at rare diseases, childhood disorders, hereditary [diseases], and cancers as a common platform to work from," De Mill said.

"The initial focus is on rare disease, where whole-exome sequencing shines," De Mill said. "We are looking to the future to roll out hereditary panels, cancer panels, and other disease states to make it simple so that our end users — either laboratories or hospitals — can work from the same platform with either somatic or germline mutations."

Another area of interest is prenatal care, according to Trabelsi.